Data Wrangling Project: Mental Health¶

Importing necessary libraries¶

In [ ]:
import pandas as pd
import seaborn as sns
import numpy as np
import math
import matplotlib.pyplot as plt
import matplotlib.colors as colors
import geopandas as gpd
import datetime

Reading in boundary data for New York City, NY¶

In [72]:
newyorkborough = gpd.read_file('geo_export_4bf61b73-3f85-41ef-9ccc-c01719c9f685.shp')
newyorkborough
newyorkborough = newyorkborough.astype({'boro_code':'int64'})
newyorkborough
Out[72]:
boro_code boro_name shape_area shape_leng geometry
0 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ...
1 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ...
2 1 Manhattan 6.365208e+08 357564.316391 MULTIPOLYGON (((-74.01093 40.68449, -74.01193 ...
3 3 Brooklyn 1.934143e+09 728197.541089 MULTIPOLYGON (((-73.86327 40.58388, -73.86381 ...
4 4 Queens 3.041419e+09 888199.731579 MULTIPOLYGON (((-73.82645 40.59053, -73.82642 ...

Reading in FACETS data for New York City, NY¶

In [77]:
facets_ = pd.read_csv('MASTER_FACETS_DATA_AllColumns.csv')
display(facets_.head())

facets_.shape
FIPS Year Urban Total_Population P_WH P_AA P_AI P_AS P_NH P_OR ... Walkscore Walkscore_Percentile Tob_Retailer_Per_1000 Felony_Per_1000 GINI SVI_Themes_Total SVI_Flags Housing_Violations_Per_1000 Voter Turnout Turnout Quartile
0 36005000100 2016 1.0 8430 0.14 0.58 0.0 0.02 0.0 0.24 ... 0.0000 0.0 0.0 40.0 NaN 0.1104 3.0 NaN 0.19 1.0
1 36005000200 2016 1.0 5095 0.32 0.28 0.0 0.06 0.0 0.30 ... -2.9096 1.0 0.0 8.0 0.4024 0.7719 1.0 135.0 0.19 1.0
2 36005000400 2016 1.0 5572 0.30 0.36 0.0 0.01 0.0 0.32 ... -2.7511 1.0 1.0 10.0 0.3659 0.6937 2.0 145.0 0.19 1.0
3 36005001600 2016 1.0 5412 0.34 0.33 0.0 0.00 0.0 0.31 ... -2.2716 2.0 1.0 9.0 0.4898 0.9451 2.0 31.0 0.20 1.0
4 36005001900 2016 1.0 2569 0.32 0.32 0.0 0.02 0.0 0.33 ... 0.2115 3.0 1.0 43.0 0.4420 0.9554 6.0 178.0 0.19 1.0

5 rows × 49 columns

Out[77]:
(6501, 49)

116 FACETS census tracts were removed with the population of 0

In [84]:
facets = facets_.drop(facets_[(facets_.Total_Population == 0)].index) # Dropped 116 Census Tracts.
print(facets.shape)

#any0 = facets[(facets.Total_Population == 0)]
#print(any0)

facets

(6385, 49)
Out[84]:
FIPS Year Urban Total_Population P_WH P_AA P_AI P_AS P_NH P_OR ... Walkscore Walkscore_Percentile Tob_Retailer_Per_1000 Felony_Per_1000 GINI SVI_Themes_Total SVI_Flags Housing_Violations_Per_1000 Voter Turnout Turnout Quartile
0 36005000100 2016 1.0 8430 0.14 0.58 0.000 0.02 0.0 0.24 ... 0.0000 0.0 0.0 40.0 NaN 0.1104 3.0 NaN 0.19 1.0
1 36005000200 2016 1.0 5095 0.32 0.28 0.000 0.06 0.0 0.30 ... -2.9096 1.0 0.0 8.0 0.4024 0.7719 1.0 135.0000 0.19 1.0
2 36005000400 2016 1.0 5572 0.30 0.36 0.000 0.01 0.0 0.32 ... -2.7511 1.0 1.0 10.0 0.3659 0.6937 2.0 145.0000 0.19 1.0
3 36005001600 2016 1.0 5412 0.34 0.33 0.000 0.00 0.0 0.31 ... -2.2716 2.0 1.0 9.0 0.4898 0.9451 2.0 31.0000 0.20 1.0
4 36005001900 2016 1.0 2569 0.32 0.32 0.000 0.02 0.0 0.33 ... 0.2115 3.0 1.0 43.0 0.4420 0.9554 6.0 178.0000 0.19 1.0
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
6495 36085030301 2018 1.0 4897 0.59 0.13 0.000 0.19 0.0 0.06 ... -2.3185 1.0 1.0 17.0 0.4005 0.5165 0.0 0.5237 0.30 4.0
6496 36085030302 2018 1.0 6142 0.54 0.19 0.000 0.14 0.0 0.08 ... -1.8722 2.0 2.0 9.0 0.4038 0.4670 0.0 1.3825 0.30 4.0
6497 36085031901 2018 1.0 2564 0.25 0.61 0.000 0.07 0.0 0.07 ... -2.2547 2.0 0.0 35.0 0.5083 0.9086 4.0 1.1520 0.30 4.0
6498 36085031902 2018 1.0 4480 0.30 0.61 0.009 0.01 0.0 0.05 ... -3.8478 1.0 1.0 15.0 0.5671 0.8321 2.0 12.0023 0.30 4.0
6499 36085032300 2018 1.0 1218 0.41 0.45 0.001 0.00 0.0 0.13 ... NaN NaN 3.0 77.0 0.5786 0.6060 1.0 1.0683 0.30 4.0

6385 rows × 49 columns

Creating new columns for the value of total felony and total housing violations by working the per 1000 equation backwards to solve for total felony and total housing violations (HV) in the FIPS location:

total felony/HV = (felonyorHV_per_1000/1000) * total population

In [4]:
facets['fel_one'] = facets['Felony_Per_1000'].div(1000).multiply(facets['Total_Population']) 
facets.head()
# (Total Felonies/Total Population) * 1000 = Felony Per 1000

facets['house_one'] = facets['Housing_Violations_Per_1000'].div(1000).multiply(facets['Total_Population']) 
# (Total Housing Violations/Total Population) * 1000 = Housing Violations Per 1000

print(facets.shape)
facets.head()

(6385, 51)
Out[4]:
FIPS Year Urban Total_Population P_WH P_AA P_AI P_AS P_NH P_OR ... Tob_Retailer_Per_1000 Felony_Per_1000 GINI SVI_Themes_Total SVI_Flags Housing_Violations_Per_1000 Voter Turnout Turnout Quartile fel_one house_one
0 36005000100 2016 1.0 8430 0.14 0.58 0.0 0.02 0.0 0.24 ... 0.0 40.0 NaN 0.1104 3.0 NaN 0.19 1.0 337.200 NaN
1 36005000200 2016 1.0 5095 0.32 0.28 0.0 0.06 0.0 0.30 ... 0.0 8.0 0.4024 0.7719 1.0 135.0 0.19 1.0 40.760 687.825
2 36005000400 2016 1.0 5572 0.30 0.36 0.0 0.01 0.0 0.32 ... 1.0 10.0 0.3659 0.6937 2.0 145.0 0.19 1.0 55.720 807.940
3 36005001600 2016 1.0 5412 0.34 0.33 0.0 0.00 0.0 0.31 ... 1.0 9.0 0.4898 0.9451 2.0 31.0 0.20 1.0 48.708 167.772
4 36005001900 2016 1.0 2569 0.32 0.32 0.0 0.02 0.0 0.33 ... 1.0 43.0 0.4420 0.9554 6.0 178.0 0.19 1.0 110.467 457.282

5 rows × 51 columns

Removing any addition symbols and commas from the median household income variable. All 250,000+ values will be considered 250,000.

In [5]:
facets[facets.MED_HH_Income == "250,000+"]
facets.loc[facets.MED_HH_Income == "250,000+", "MED_HH_Income"]="250000"
facets[facets.MED_HH_Income == "250000"]

facets["MED_HH_Income"] = facets.MED_HH_Income.astype(str).astype(float)
#print(facets.dtypes)

Adding the county and city names for each FIPS location using their county FIPS identifiers.

In [6]:
county_dict = {'Bronx County':36005,'Kings County': 36047, 
               'New York County': 36061, 'Queens County':36081, 'Richmond County': 36085 }
County_df = pd.DataFrame.from_dict(county_dict,orient='index',columns=["FIPS_code"])
county = ['Bronx County','Kings County', 'New York County', 'Queens County', 'Richmond County']
city = ['Bronx','Brooklyn','Manhattan','Queens','Staten Island']
boro_code = ['2','3','1','4','5']
County_df['County'] = county
County_df['City'] = city
County_df['boro_code'] = boro_code
County_df.head()

fips_list = facets['FIPS'].tolist()
#fips_list
fips_str = map(str,fips_list)
fips_str_list = list(fips_str)

def five_digits(num):
    return int(str(num[:5]))

fips_result = [five_digits(n) for n in fips_str_list]
#print(fips_result)

facets['FIPS_code'] = fips_result
facets_county = facets.merge(County_df, on=['FIPS_code'])
facets_county
Out[6]:
FIPS Year Urban Total_Population P_WH P_AA P_AI P_AS P_NH P_OR ... SVI_Flags Housing_Violations_Per_1000 Voter Turnout Turnout Quartile fel_one house_one FIPS_code County City boro_code
0 36005000100 2016 1.0 8430 0.14 0.58 0.000 0.02 0.0 0.24 ... 3.0 NaN 0.19 1.0 337.200 NaN 36005 Bronx County Bronx 2
1 36005000200 2016 1.0 5095 0.32 0.28 0.000 0.06 0.0 0.30 ... 1.0 135.0000 0.19 1.0 40.760 687.825000 36005 Bronx County Bronx 2
2 36005000400 2016 1.0 5572 0.30 0.36 0.000 0.01 0.0 0.32 ... 2.0 145.0000 0.19 1.0 55.720 807.940000 36005 Bronx County Bronx 2
3 36005001600 2016 1.0 5412 0.34 0.33 0.000 0.00 0.0 0.31 ... 2.0 31.0000 0.20 1.0 48.708 167.772000 36005 Bronx County Bronx 2
4 36005001900 2016 1.0 2569 0.32 0.32 0.000 0.02 0.0 0.33 ... 6.0 178.0000 0.19 1.0 110.467 457.282000 36005 Bronx County Bronx 2
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
6380 36085030301 2018 1.0 4897 0.59 0.13 0.000 0.19 0.0 0.06 ... 0.0 0.5237 0.30 4.0 83.249 2.564559 36085 Richmond County Staten Island 5
6381 36085030302 2018 1.0 6142 0.54 0.19 0.000 0.14 0.0 0.08 ... 0.0 1.3825 0.30 4.0 55.278 8.491315 36085 Richmond County Staten Island 5
6382 36085031901 2018 1.0 2564 0.25 0.61 0.000 0.07 0.0 0.07 ... 4.0 1.1520 0.30 4.0 89.740 2.953728 36085 Richmond County Staten Island 5
6383 36085031902 2018 1.0 4480 0.30 0.61 0.009 0.01 0.0 0.05 ... 2.0 12.0023 0.30 4.0 67.200 53.770304 36085 Richmond County Staten Island 5
6384 36085032300 2018 1.0 1218 0.41 0.45 0.001 0.00 0.0 0.13 ... 1.0 1.0683 0.30 4.0 93.786 1.301189 36085 Richmond County Staten Island 5

6385 rows × 55 columns

Obtaining a copy of the dataframe for my specific factor, Mental Health, with the applicable variables. Assessing the variable types.

In [85]:
mental_health = facets_county.iloc[:,[0,1,3,24,25,42,43,46,49,50,53,54]].copy()

datatypes = mental_health.dtypes
#print(datatypes)
mental_health["boro_code"] = mental_health.boro_code.astype(str).astype('int64')
print(datatypes)

mental_health.shape

FIPS                             int64
Year                             int64
Total_Population                 int64
Poverty_Rate                   float64
MED_HH_Income                  float64
Felony_Per_1000                float64
GINI                           float64
Housing_Violations_Per_1000    float64
fel_one                        float64
house_one                      float64
City                            object
boro_code                       object
dtype: object
Out[85]:
(6385, 12)

Performing descriptive statistics using the describe function. We can see that many of the variables have a positive skew based on the mean and median values alone: total population, poverty rate, and median house hold income. Felony per 1000 and Housing Violations have a very strong positive skew. GINI is the only variable that appears to be normally distributed.

In [86]:
mental_health[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000"]].describe()
Out[86]:
Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000
count 6385.000000 6375.000000 6316.000000 6369.000000 6337.000000 5727.000000
mean 3953.577604 0.195319 59678.697277 51.833726 0.453825 71.756139
std 2196.715908 0.131834 29392.449593 550.735166 0.064689 178.954187
min 1.000000 0.000000 9327.000000 0.000000 0.125100 0.000000
25% 2423.000000 0.090000 39468.000000 9.000000 0.409600 0.102200
50% 3590.000000 0.170000 55181.500000 17.000000 0.449000 0.420500
75% 4995.000000 0.270000 74099.500000 43.000000 0.491000 40.000000
max 29256.000000 1.000000 250000.000000 41000.000000 0.690500 2714.000000

Assessing the occurrence of Null values over the whole dataset, then by year.

In [9]:
null_mental = mental_health.isnull().sum()
print(null_mental)

len(mental_health.index) # Highest Null values is at 10% (Housing_Violations_Per_1000), no variables need to be removed.

FIPS                             0
Year                             0
Total_Population                 0
Poverty_Rate                    10
MED_HH_Income                   69
Felony_Per_1000                 16
GINI                            48
Housing_Violations_Per_1000    658
fel_one                         16
house_one                      658
City                             0
boro_code                        0
dtype: int64
Out[9]:
6385
In [10]:
year16 = mental_health[mental_health.Year == 2016]
year16.isnull().sum()
Out[10]:
FIPS                             0
Year                             0
Total_Population                 0
Poverty_Rate                     4
MED_HH_Income                   19
Felony_Per_1000                 14
GINI                            18
Housing_Violations_Per_1000    532
fel_one                         14
house_one                      532
City                             0
boro_code                        0
dtype: int64
In [11]:
year17 = mental_health[mental_health.Year == 2017]
year17.isnull().sum()
Out[11]:
FIPS                            0
Year                            0
Total_Population                0
Poverty_Rate                    3
MED_HH_Income                  25
Felony_Per_1000                 2
GINI                           15
Housing_Violations_Per_1000    63
fel_one                         2
house_one                      63
City                            0
boro_code                       0
dtype: int64
In [12]:
year18 = mental_health[mental_health.Year == 2018]
year18.isnull().sum()
Out[12]:
FIPS                            0
Year                            0
Total_Population                0
Poverty_Rate                    3
MED_HH_Income                  25
Felony_Per_1000                 0
GINI                           15
Housing_Violations_Per_1000    63
fel_one                         0
house_one                      63
City                            0
boro_code                       0
dtype: int64

FIPS¶

Assessing the FIPS locations and if they have recordings for each year. Removed those missing recordings for each year.

In [13]:
fips_occ = pd.DataFrame(mental_health['FIPS'].value_counts())
fips_occ

fips_occ.value_counts()

fips_occ.tail(5)
Out[13]:
count
FIPS
36005002400 2
36061021703 2
36005016300 1
36081038302 1
36005017100 1
In [14]:
fp = mental_health[(mental_health.FIPS == 36005002400) | (mental_health.FIPS == 36061021703) |
                        (mental_health.FIPS == 36005016300) | (mental_health.FIPS == 36081038302) | (mental_health.FIPS == 36005017100)]
display(fp)
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
96 36005016300 2016 77 0.57 NaN 571.0 NaN NaN 43.967 NaN Bronx 2
102 36005017100 2016 1 0.00 NaN NaN NaN NaN NaN NaN Bronx 2
344 36005002400 2017 150 0.00 NaN 1.0 NaN NaN 0.150 NaN Bronx 2
680 36005002400 2018 123 0.00 NaN 24.0 NaN NaN 2.952 NaN Bronx 2
3764 36061021703 2017 8 0.00 NaN 80.0 NaN 0.0 0.640 0.0 Manhattan 1
4047 36061021703 2018 8 0.00 NaN 11125.0 NaN 0.0 89.000 0.0 Manhattan 1
4358 36081038302 2016 44 NaN NaN NaN NaN NaN NaN NaN Queens 4
In [15]:
fips_pop = mental_health[(mental_health.FIPS == 36005002400) | (mental_health.FIPS == 36061021703) |
                        (mental_health.FIPS == 36005016300) | (mental_health.FIPS == 36081038302) | (mental_health.FIPS == 36005017100)].index

mental_health.drop(fips_pop, inplace=True)

Total Population¶

Assessing the normality of total population since it is needed for evaluating felony per 1000 and housing per 1000. Has a positive skew.

In [16]:
totalpopulation_hist = sns.histplot(data=mental_health,x='Total_Population', color='red').set(title="Total Population Histogram")
plt.show()

totalpopulation_box = sns.boxplot(data=mental_health,y='Total_Population', color='red').set(title="Total Population Boxplot")
plt.show()

Assessing the outliers noticed in the boxplot. All of them are consistent between the years. I believe this to be true values and not an error. Leaving values as is.

In [17]:
pd.options.mode.chained_assignment = None 

pop_pop = mental_health[(mental_health.Total_Population >=14800)].sort_values(['FIPS'])
pop_pop
Out[17]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
332 36005046201 2016 28827 0.08 44856.0 10.0 0.4250 2714.0000 288.270 78236.478000 Bronx 2
668 36005046201 2017 28926 0.07 43205.0 334.0 0.4348 0.3077 9661.284 8.900530 Bronx 2
1004 36005046201 2018 29256 0.07 45770.0 12.0 0.4348 0.3077 351.072 9.002071 Bronx 2
3304 36061004400 2016 16735 0.12 92267.0 2.0 0.4407 1.0000 33.470 16.735000 Manhattan 1
3586 36061004400 2017 16200 0.13 93010.0 54.0 0.4449 0.0241 874.800 0.390420 Manhattan 1
3869 36061004400 2018 16039 0.10 94000.0 3.0 0.4449 0.0241 48.117 0.386540 Manhattan 1
3395 36061013600 2016 15641 0.04 107343.0 4.0 0.5686 69.0000 62.564 1079.229000 Manhattan 1
3677 36061013600 2017 15880 0.03 108587.0 109.0 0.5569 0.7484 1730.920 11.884592 Manhattan 1
3960 36061013600 2018 15234 0.04 101554.0 6.0 0.5569 0.7484 91.404 11.401126 Manhattan 1
3416 36061015400 2016 14969 0.08 83767.0 5.0 0.4782 66.0000 74.845 987.954000 Manhattan 1
3981 36061015400 2018 15502 0.07 82026.0 7.0 0.4687 0.6481 108.514 10.046846 Manhattan 1
3512 36061024500 2016 15899 0.32 34564.0 9.0 0.4199 952.0000 143.091 15135.848000 Manhattan 1
3795 36061024500 2017 16231 0.31 35394.0 259.0 0.4245 7.0808 4203.829 114.928465 Manhattan 1
4078 36061024500 2018 15642 0.29 38321.0 13.0 0.4245 7.0808 203.346 110.757874 Manhattan 1
In [18]:
concern_tpop = set(pop_pop['FIPS'])
#print(concern_tpop)

#ctp = list(pop_pop['FIPS'])
#ctp

out_tpop = mental_health['FIPS'].isin(concern_tpop)
consistency_tpop = mental_health[out_tpop]

consistency_tpop['Year'] = pd.to_datetime(consistency_tpop['Year'], format='%Y')
consistency_tpop.set_index('Year',inplace=True)

consistency_tpop.groupby('FIPS')['Total_Population'].plot(legend=True,marker='o',grid=True)
plt.title('Total Population Outlier Consistency')
plt.ylabel('Total Population')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
#consistency_tpop
plt.show()

Poverty Rate¶

Assessing Poverty Rate for outliers and distribution. There is a positive skew on the data and there are some outliers. They are consistent between the years and do not indicate transcription errors.

In [19]:
povertyrate_hist  = sns.histplot(data=mental_health, x='Poverty_Rate', color='blue').set(title="Poverty Rate Histogram")
plt.show()

povertyrate_box  = sns.boxplot(data=mental_health, y='Poverty_Rate', color='blue').set(title="Poverty Rate Boxplot")
plt.show()
In [20]:
pov_pop = mental_health[(mental_health.Poverty_Rate >=0.8)].sort_values(['FIPS'])
pov_pop
Out[20]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
225 36005031900 2016 647 0.93 NaN 6.0 NaN NaN 3.882 NaN Bronx 2
561 36005031900 2017 645 0.84 NaN 8.0 NaN 0.0000 5.160 0.000000 Bronx 2
3401 36061014300 2016 4 1.00 NaN NaN NaN NaN NaN NaN Manhattan 1
3683 36061014300 2017 5 1.00 NaN 230.0 NaN NaN 1.150 NaN Manhattan 1
3507 36061024000 2016 1580 0.85 NaN 44.0 NaN NaN 69.520 NaN Manhattan 1
3790 36061024000 2017 1688 0.88 NaN 155.0 NaN 0.0037 261.640 0.006246 Manhattan 1
4073 36061024000 2018 1742 0.90 NaN 52.0 NaN 0.0037 90.584 0.006445 Manhattan 1
In [21]:
concern_pov = set(pov_pop['FIPS'])
#print(concern_pov)

#cpv = list(pov_pop['FIPS'])
#cpv

out_pov = mental_health['FIPS'].isin(concern_pov)
consistency_pov = mental_health[out_pov]

consistency_pov['Year'] = pd.to_datetime(consistency_pov['Year'], format='%Y')
consistency_pov.set_index('Year',inplace=True)


consistency_pov.groupby('FIPS')['Poverty_Rate'].plot(legend=True,marker='o',grid=True, figsize=(5,5))
plt.title('Poverty Rate Outlier Consistency')
plt.ylabel('Poverty Rate')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
#consistency_pov
plt.show()

consistency_pov.groupby('FIPS')['Total_Population'].plot(legend=True,marker='o',grid=True, figsize=(5,5))
plt.title('Total Population Compare to Poverty Rate Outliers')
plt.ylabel('Total Population')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
plt.show()
In [22]:
mental_health.loc[(mental_health['FIPS'] == 36061014300) | (mental_health['FIPS'] == 36005031900)]
Out[22]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
225 36005031900 2016 647 0.93 NaN 6.0 NaN NaN 3.882 NaN Bronx 2
561 36005031900 2017 645 0.84 NaN 8.0 NaN 0.0 5.160 0.0 Bronx 2
897 36005031900 2018 654 0.72 NaN 8.0 NaN 0.0 5.232 0.0 Bronx 2
3401 36061014300 2016 4 1.00 NaN NaN NaN NaN NaN NaN Manhattan 1
3683 36061014300 2017 5 1.00 NaN 230.0 NaN NaN 1.150 NaN Manhattan 1
3966 36061014300 2018 5 0.75 NaN 41000.0 NaN NaN 205.000 NaN Manhattan 1
In [23]:
mental_health['Poverty_Rate'] = mental_health['Poverty_Rate'].fillna(mental_health['Poverty_Rate'].median())

Median Household Income¶

Assessing median household income. There is a slight positive skew and several outliers. The outliers are consistent between the years and don't indicate to being a data entry error.

In [24]:
medhh_hist = sns.histplot(data=mental_health,x='MED_HH_Income', color='green').set(title="Median Household Income Histogram")
plt.show()

medhh_box = sns.boxplot(data=mental_health,y='MED_HH_Income', color='green').set(title="Median Household Income Boxplot")
plt.show()
In [25]:
inc_pop = mental_health[(mental_health.MED_HH_Income >=200000)].sort_values(['FIPS'])
inc_pop
Out[25]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
1021 36047002100 2016 3798 0.07 204205.0 27.0 0.3995 NaN 102.546 NaN Brooklyn 3
1772 36047002100 2017 4295 0.08 208036.0 165.0 0.3931 0.0115 708.675 0.049392 Brooklyn 3
2523 36047002100 2018 4628 0.07 211071.0 35.0 0.3931 0.0115 161.980 0.053222 Brooklyn 3
3844 36061002100 2018 6488 0.03 250000.0 23.0 0.5001 0.0464 149.224 0.301043 Manhattan 1
3561 36061002100 2017 6735 0.03 250000.0 174.0 0.5001 0.0464 1171.890 0.312504 Manhattan 1
3279 36061002100 2016 6506 0.02 232266.0 18.0 0.5101 44.0000 117.108 286.264000 Manhattan 1
3575 36061003300 2017 5005 0.04 250000.0 226.0 0.5818 0.0241 1131.130 0.120620 Manhattan 1
3858 36061003300 2018 5030 0.05 207460.0 35.0 0.5818 0.0241 176.050 0.121223 Manhattan 1
3293 36061003300 2016 5262 0.06 219554.0 26.0 0.5863 13.0000 136.812 68.406000 Manhattan 1
3892 36061006900 2018 2442 0.06 209167.0 43.0 0.5195 0.0371 105.006 0.090598 Manhattan 1
3369 36061011201 2016 843 0.03 225250.0 262.0 0.4655 41.0000 220.866 34.563000 Manhattan 1
3389 36061013000 2016 3693 0.08 217070.0 9.0 0.5847 60.0000 33.237 221.580000 Manhattan 1
3671 36061013000 2017 3277 0.07 244375.0 51.0 0.5668 0.0687 167.127 0.225130 Manhattan 1
3400 36061014200 2016 4288 0.08 225208.0 6.0 0.6038 12.0000 25.728 51.456000 Manhattan 1
3965 36061014200 2018 4186 0.05 235083.0 8.0 0.6043 0.0891 33.488 0.372973 Manhattan 1
3682 36061014200 2017 4467 0.06 235455.0 55.0 0.6043 0.0891 245.685 0.398010 Manhattan 1
3706 36061016001 2017 3861 0.02 222222.0 19.0 0.5450 0.0353 73.359 0.136293 Manhattan 1
3989 36061016001 2018 3918 0.02 233529.0 4.0 0.5450 0.0353 15.672 0.138305 Manhattan 1
In [26]:
concern_inc = set(inc_pop['FIPS'])
#print(concern_inc)

#cinc = list(inc_pop['FIPS'])
#cinc

out_inc = mental_health['FIPS'].isin(concern_inc)
consistency_inc = mental_health[out_inc]

consistency_inc['Year'] = pd.to_datetime(consistency_inc['Year'], format='%Y')
consistency_inc.set_index('Year',inplace=True)

consistency_inc.groupby('FIPS')['MED_HH_Income'].plot(legend=True,marker='o',grid=True,figsize=(10,10))
plt.title('Median Household Income Outlier Consistency')
plt.ylabel('Median Household Income')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
#consistency_inc
plt.show()

consistency_inc.groupby('FIPS')['Total_Population'].plot(legend=True,marker='o',grid=True,figsize=(10,10))
plt.title('Total Population Compare to Median Household Income Outliers')
plt.ylabel('Total Population')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
plt.show()
In [27]:
mental_health.loc[mental_health['FIPS'] == 36061011201]
Out[27]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
3369 36061011201 2016 843 0.03 225250.0 262.0 0.4655 41.0 220.866 34.563 Manhattan 1
3651 36061011201 2017 792 0.07 138250.0 232.0 0.5330 0.0 183.744 0.000 Manhattan 1
3934 36061011201 2018 837 0.07 132788.0 287.0 0.5330 0.0 240.219 0.000 Manhattan 1
In [28]:
mental_health['MED_HH_Income'] = mental_health['MED_HH_Income'].fillna(mental_health['MED_HH_Income'].median())

Felonies Per 1000¶

Assesesing felony per 1000. There are several outliers and it has a significant positive skew. Eight values were changed to NaN and replaced with the median value along with the original NaN values.

In [29]:
felony_box = sns.boxplot(data=mental_health, y='Felony_Per_1000', color='purple').set(title="Felony Per 1000 Box Plot")
plt.show()

felony_box_out = sns.boxplot(data=mental_health, y='Felony_Per_1000', showfliers=False, color='purple').set(title="Felony Per 1000 Box Plot (Outliers Removed)")
plt.show()

fel_box = sns.boxplot(data=facets, y='fel_one', color='purple').set(title="# of Felonies in Total Population of FIPS location")
plt.show()

fel_hist = sns.boxplot(data=facets, y='fel_one',showfliers=False, color='purple').set(title="# of Felonies in Total Population of FIPS location (Outliers removed)")
plt.show()

mental_health['Felony_Per_1000'] = mental_health['Felony_Per_1000'].fillna(mental_health['Felony_Per_1000'].median())
In [30]:
pop_fel = mental_health[(mental_health.fel_one >=10) & (mental_health.Total_Population <=800)]
pop_fel.sort_values(['FIPS'])
Out[30]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
14 36005003700 2016 199 0.50 9327.0 70.0 0.4719 NaN 13.930 NaN Bronx 2
688 36005003700 2018 279 0.39 55181.5 82.0 0.4664 0.1465 22.878 0.040873 Bronx 2
736 36005011000 2018 151 0.21 55181.5 79.0 0.3533 0.0176 11.929 0.002658 Bronx 2
199 36005027600 2016 62 0.11 54167.0 516.0 0.2626 NaN 31.992 NaN Bronx 2
871 36005027600 2018 59 0.14 60000.0 610.0 0.2929 NaN 35.990 NaN Bronx 2
204 36005028400 2016 707 0.27 96094.0 73.0 0.2624 NaN 51.611 NaN Bronx 2
540 36005028400 2017 740 0.24 55181.5 66.0 0.2923 0.0000 48.840 0.000000 Bronx 2
876 36005028400 2018 687 0.29 98482.0 95.0 0.2923 0.0000 65.265 0.000000 Bronx 2
320 36005043500 2016 55 0.00 20417.0 1745.0 0.2988 777.0000 95.975 42.735000 Bronx 2
992 36005043500 2018 62 0.13 55181.5 1839.0 0.2455 0.0059 114.018 0.000366 Bronx 2
335 36005050400 2016 54 0.00 29844.0 259.0 0.6393 NaN 13.986 NaN Bronx 2
1007 36005050400 2018 65 0.00 55181.5 246.0 0.6682 NaN 15.990 NaN Bronx 2
1016 36047001100 2016 701 0.09 103500.0 262.0 0.3045 NaN 183.662 NaN Brooklyn 3
2743 36047028501 2018 432 0.27 92308.0 95.0 0.3762 0.0627 41.040 0.027086 Brooklyn 3
1992 36047028501 2017 293 0.12 84861.0 52.0 0.3762 0.0627 15.236 0.018371 Brooklyn 3
1241 36047028501 2016 423 0.10 97083.0 54.0 0.3167 NaN 22.842 NaN Brooklyn 3
1393 36047045000 2016 507 0.31 78438.0 33.0 0.5438 104.0000 16.731 52.728000 Brooklyn 3
2895 36047045000 2018 514 0.47 45938.0 47.0 0.6045 0.0000 24.158 0.000000 Brooklyn 3
1469 36047054300 2016 347 0.11 81583.0 95.0 0.4062 NaN 32.965 NaN Brooklyn 3
2220 36047054300 2017 382 0.10 92500.0 43.0 0.3787 0.0066 16.426 0.002521 Brooklyn 3
2971 36047054300 2018 351 0.01 119250.0 125.0 0.3787 0.0066 43.875 0.002317 Brooklyn 3
3128 36047085200 2018 8 0.00 55181.5 2000.0 NaN NaN 16.000 NaN Brooklyn 3
3917 36061009400 2018 44 0.00 199167.0 2477.0 0.4154 0.0316 108.988 0.001390 Manhattan 1
3919 36061009600 2018 119 0.11 136406.0 1471.0 0.4754 0.0539 175.049 0.006414 Manhattan 1
3636 36061009600 2017 149 0.08 135313.0 188.0 0.4754 0.0539 28.012 0.008031 Manhattan 1
3354 36061009600 2016 147 0.08 91875.0 1040.0 0.5146 NaN 152.880 NaN Manhattan 1
3360 36061010200 2016 225 0.14 100500.0 497.0 0.5688 NaN 111.825 NaN Manhattan 1
3642 36061010200 2017 258 0.08 55181.5 106.0 0.5213 0.0037 27.348 0.000955 Manhattan 1
3925 36061010200 2018 180 0.12 165083.0 689.0 0.5213 0.0037 124.020 0.000666 Manhattan 1
3931 36061010900 2018 147 0.15 55181.5 4204.0 0.4169 NaN 617.988 NaN Manhattan 1
3648 36061010900 2017 186 0.17 135750.0 746.0 0.4169 NaN 138.756 NaN Manhattan 1
3366 36061010900 2016 220 0.18 132500.0 2381.0 0.4034 NaN 523.820 NaN Manhattan 1
3651 36061011201 2017 792 0.07 138250.0 232.0 0.5330 0.0000 183.744 0.000000 Manhattan 1
3935 36061011202 2018 257 0.05 187500.0 350.0 0.5871 0.0000 89.950 0.000000 Manhattan 1
3370 36061011202 2016 266 0.02 153594.0 312.0 0.6294 NaN 82.992 NaN Manhattan 1
3652 36061011202 2017 287 0.05 192857.0 82.0 0.5871 0.0000 23.534 0.000000 Manhattan 1
3372 36061011300 2016 146 0.26 81786.0 1075.0 0.5664 NaN 156.950 NaN Manhattan 1
3937 36061011300 2018 120 0.23 90938.0 1575.0 0.5721 0.0000 189.000 0.000000 Manhattan 1
3654 36061011300 2017 126 0.19 82361.0 236.0 0.5721 0.0000 29.736 0.000000 Manhattan 1
3966 36061014300 2018 5 0.75 55181.5 41000.0 NaN NaN 205.000 NaN Manhattan 1
4027 36061019701 2018 749 0.40 68092.0 48.0 0.5439 0.0000 35.952 0.000000 Manhattan 1
4103 36061029700 2018 18 0.17 55181.5 1944.0 NaN NaN 34.992 NaN Manhattan 1
5571 36081021900 2018 4 0.00 55181.5 7750.0 NaN NaN 31.000 NaN Queens 4
4376 36081042600 2016 500 0.65 55181.5 24.0 NaN NaN 12.000 NaN Queens 4
5026 36081042600 2017 512 0.63 55181.5 23.0 NaN NaN 11.776 NaN Queens 4
5676 36081042600 2018 507 0.67 55181.5 41.0 NaN NaN 20.787 NaN Queens 4
5806 36081060701 2018 15 0.00 55181.5 1200.0 NaN NaN 18.000 NaN Queens 4
4534 36081064102 2016 134 0.00 20250.0 201.0 0.4226 NaN 26.934 NaN Queens 4
5834 36081064102 2018 172 0.33 55181.5 198.0 0.3973 NaN 34.056 NaN Queens 4
6029 36081138502 2018 37 0.49 55181.5 703.0 NaN NaN 26.011 NaN Queens 4
6362 36085022800 2018 41 0.00 55181.5 927.0 0.1251 NaN 38.007 NaN Staten Island 5
In [31]:
concern_fel = set(pop_fel['FIPS'])
#print(concern_fel)

#cfel = list(pop_fel['FIPS'])
#cfel

out_fel = mental_health['FIPS'].isin(concern_fel)
consistency_fel = mental_health[out_fel]

consistency_fel['Year'] = pd.to_datetime(consistency_fel['Year'], format='%Y')
consistency_fel.set_index('Year',inplace=True)


consistency_fel.groupby('FIPS')['Felony_Per_1000'].plot(legend=True,marker='o',grid=True,figsize=(10,10))
plt.title('Felony Per 1000 Outlier Consistency')
plt.ylabel('Felony Per 1000')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
#consistency_fel
plt.show()

consistency_fel.groupby('FIPS')['Total_Population'].plot(legend=True,marker='o',grid=True,figsize=(10,10))
plt.title('Total Population Compare to Felony Per 1000 Outliers')
plt.ylabel('Total Population')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
plt.show()
In [32]:
problems = mental_health[(mental_health.FIPS ==36005043500) | (mental_health.FIPS ==36047085200) |
                        (mental_health.FIPS ==36061010900) | (mental_health.FIPS ==36061009400) |
                        (mental_health.FIPS ==36061010900) | (mental_health.FIPS ==36061014300) | 
                         (mental_health.FIPS ==36081021900) ].sort_values('FIPS')
problems.iloc[:,[0,2,5,8]]
Out[32]:
FIPS Total_Population Felony_Per_1000 fel_one
320 36005043500 55 1745.0 95.975
656 36005043500 52 182.0 9.464
992 36005043500 62 1839.0 114.018
1626 36047085200 8 17.0 NaN
2377 36047085200 8 12.0 0.096
3128 36047085200 8 2000.0 16.000
3352 36061009400 33 17.0 NaN
3634 36061009400 15 116.0 1.740
3917 36061009400 44 2477.0 108.988
3366 36061010900 220 2381.0 523.820
3648 36061010900 186 746.0 138.756
3931 36061010900 147 4204.0 617.988
3966 36061014300 5 41000.0 205.000
3401 36061014300 4 17.0 NaN
3683 36061014300 5 230.0 1.150
4921 36081021900 4 31.0 0.124
4270 36081021900 4 17.0 NaN
5571 36081021900 4 7750.0 31.000
In [33]:
#problems['Year'] = pd.to_datetime(problems['Year'], format='%Y')
#problems.set_index('Year',inplace=True)

problems.groupby('FIPS')['Felony_Per_1000'].plot(legend=True,marker='o',grid=True,figsize=(5,5))
plt.title('Felony Per 1000')
plt.ylabel('Felony Per 1000')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
plt.show()


major_problem =  mental_health[mental_health.FIPS == 36061014300]

major_problem['Year'] = pd.to_datetime(major_problem['Year'], format='%Y')
major_problem.set_index('Year',inplace=True)

major_problem.groupby('FIPS')['Felony_Per_1000'].plot(legend=True,marker='o',grid=True,figsize=(5,5))
plt.title('Felony Per 1000 FIPS 36061014300')
plt.ylabel('Felony Per 1000')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
plt.show()


minor_problems =  mental_health[(mental_health.FIPS ==36005043500) | (mental_health.FIPS ==36047085200) |
                        (mental_health.FIPS ==36061010900) | (mental_health.FIPS ==36061009400) |
                        (mental_health.FIPS ==36061010900) | (mental_health.FIPS ==36081021900)]

minor_problems['Year'] = pd.to_datetime(minor_problems['Year'], format='%Y')
minor_problems.set_index('Year',inplace=True)                         

minor_problems.groupby('FIPS')['Felony_Per_1000'].plot(legend=True,marker='o',grid=True,figsize=(5,5))
plt.title('Felony Per 1000 Minor Issues')
plt.ylabel('Felony Per 1000')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
plt.show()
In [34]:
mental_health.at[320,'Felony_Per_1000']=np.nan #36005043500 2018 55/95
mental_health.at[992,'Felony_Per_1000']=np.nan #36005043500 2018 62/114
mental_health.at[3128,'Felony_Per_1000']=np.nan #36047085200 2018 8/16
mental_health.at[3366,'Felony_Per_1000']=np.nan #36061010900 2016 220/523
mental_health.at[3917,'Felony_Per_1000']=np.nan #36061009400 2018 44/108
mental_health.at[3931,'Felony_Per_1000']=np.nan #36061010900 2018 147/617
mental_health.at[3966,'Felony_Per_1000']=np.nan #36061014300 2015 5/205
mental_health.at[5571,'Felony_Per_1000']=np.nan #36081021900 2018 4/31


sns.boxplot(mental_health,y='Felony_Per_1000', color='purple')
plt.show()
sns.histplot(mental_health,x='Felony_Per_1000',bins=10, color='purple')
plt.show()
In [35]:
mental_health['Felony_Per_1000'] = mental_health['Felony_Per_1000'].fillna(mental_health['Felony_Per_1000'].median())

GINI Index¶

GINI Index had a normal distribution with minimal outliers. The outliers are consistent through the years.

In [36]:
gini_hist = sns.histplot(data=mental_health, x='GINI', color='yellow').set(title="GINI Index Histogram")
plt.show()

gini_box = sns.boxplot(data=mental_health, y='GINI', color='yellow').set(title="GINI Index Boxplot")
plt.show()
In [37]:
gini_pop = mental_health[(mental_health.GINI <=0.2) | (mental_health.GINI >=0.65)].sort_values(['FIPS'])
gini_pop
Out[37]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
116 36005019300 2016 5301 0.55 17268.0 17.0 0.6509 335.0000 90.117 1775.835000 Bronx 2
671 36005050400 2017 63 0.00 55181.5 2.0 0.6682 NaN 0.126 NaN Bronx 2
1007 36005050400 2018 65 0.00 55181.5 246.0 0.6682 NaN 15.990 NaN Bronx 2
1043 36047005201 2016 1716 0.32 17129.0 2.0 0.6679 NaN 3.432 NaN Brooklyn 3
1313 36047036502 2016 1101 0.40 37708.0 28.0 0.6536 966.0000 30.828 1063.566000 Brooklyn 3
1357 36047041402 2016 1735 0.22 31063.0 5.0 0.6553 286.0000 8.675 496.210000 Brooklyn 3
2108 36047041402 2017 1611 0.23 28884.0 17.0 0.6806 0.1055 27.387 0.169961 Brooklyn 3
2859 36047041402 2018 1765 0.17 38375.0 7.0 0.6806 0.1055 12.355 0.186207 Brooklyn 3
3577 36061003601 2017 3238 0.33 44286.0 85.0 0.6639 0.1616 275.230 0.523261 Manhattan 1
3860 36061003601 2018 3401 0.24 48990.0 26.0 0.6639 0.1616 88.426 0.549602 Manhattan 1
3916 36061009300 2018 10012 0.18 74828.0 17.0 0.6516 0.2080 170.204 2.082496 Manhattan 1
3633 36061009300 2017 10650 0.25 76058.0 192.0 0.6516 0.2080 2044.800 2.215200 Manhattan 1
3944 36061011900 2018 996 0.50 12903.0 569.0 0.6503 0.1226 566.724 0.122110 Manhattan 1
3661 36061011900 2017 1068 0.56 11671.0 618.0 0.6503 0.1226 660.024 0.130937 Manhattan 1
3994 36061016400 2018 7260 0.27 32347.0 17.0 0.6609 0.3008 123.420 2.183808 Manhattan 1
3711 36061016400 2017 6901 0.24 33493.0 118.0 0.6609 0.3008 814.318 2.075821 Manhattan 1
3429 36061016400 2016 6722 0.22 33792.0 10.0 0.6580 55.0000 67.220 369.710000 Manhattan 1
3433 36061016800 2016 4974 0.38 26069.0 14.0 0.6905 34.0000 69.636 169.116000 Manhattan 1
3998 36061016800 2018 4681 0.37 31944.0 21.0 0.6721 0.1838 98.301 0.860368 Manhattan 1
3715 36061016800 2017 4693 0.38 26136.0 124.0 0.6721 0.1838 581.932 0.862573 Manhattan 1
3721 36061017401 2017 4127 0.37 21435.0 96.0 0.6580 0.3398 396.192 1.402355 Manhattan 1
3439 36061017401 2016 4330 0.35 20607.0 9.0 0.6505 112.0000 38.970 484.960000 Manhattan 1
4004 36061017401 2018 4508 0.44 20288.0 16.0 0.6580 0.3398 72.128 1.531818 Manhattan 1
4511 36081061302 2016 14 0.00 55181.5 17.0 0.1862 NaN NaN NaN Queens 4
5161 36081061302 2017 15 0.00 55181.5 17.0 0.1680 NaN NaN NaN Queens 4
5811 36081061302 2018 14 0.00 55181.5 0.0 0.1680 NaN 0.000 NaN Queens 4
5367 36081128300 2017 29 0.00 32188.0 17.0 0.1668 NaN 0.493 NaN Queens 4
6017 36081128300 2018 47 0.00 43889.0 106.0 0.1668 NaN 4.982 NaN Queens 4
6254 36085022800 2017 39 0.00 55181.5 38.0 0.1251 NaN 1.482 NaN Staten Island 5
6362 36085022800 2018 41 0.00 55181.5 927.0 0.1251 NaN 38.007 NaN Staten Island 5
In [38]:
concern_gini = set(gini_pop['FIPS'])
#print(concern_gini)

#cgini = list(gini_pop['FIPS'])
#cgini

out_gini = mental_health['FIPS'].isin(concern_gini)
consistency_gini = mental_health[out_gini]

consistency_gini['Year'] = pd.to_datetime(consistency_gini['Year'], format='%Y')
consistency_gini.set_index('Year',inplace=True)


consistency_gini.groupby('FIPS')['GINI'].plot(legend=True,marker='o',grid=True,figsize=(10,10))
plt.title('GINI Index Outlier Consistency')
plt.ylabel('GINI Index')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
#consistency_gini
plt.show()

consistency_gini.groupby('FIPS')['Total_Population'].plot(legend=True,marker='o',grid=True,figsize=(10,10))
plt.title('Total Population Compare to GINI Index Outliers')
plt.ylabel('Total Population')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
plt.show()
In [39]:
gini_concern = mental_health[mental_health.FIPS == 36085022800]
gini_concern
Out[39]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
6146 36085022800 2016 231 0.0 55181.5 17.0 NaN NaN NaN NaN Staten Island 5
6254 36085022800 2017 39 0.0 55181.5 38.0 0.1251 NaN 1.482 NaN Staten Island 5
6362 36085022800 2018 41 0.0 55181.5 927.0 0.1251 NaN 38.007 NaN Staten Island 5
In [40]:
mental_health['GINI'] = mental_health['GINI'].fillna(mental_health['GINI'].median())

Housing Violations Per 1000¶

Housing violations per 1000 had significant outliers that were assessed as being correct. The year 2016 had an increase for housing violations for every FIPS location, so I didn't want to incorrectly remove any values and impute them.

In [41]:
housing_box = sns.boxplot(data=mental_health, y='Housing_Violations_Per_1000', color='orange').set(title="Housing Violations Per 1000 Boxplot")
plt.show()

housing_box = sns.boxplot(data=mental_health, y='Housing_Violations_Per_1000', showfliers=False, color='orange').set(title="Housing Violations Per 1000 Boxplot (Outliers Removed)")
plt.show()

house_box = sns.boxplot(data=facets, y='house_one', color='orange').set(title="# of Housing Violations in Total Population of FIPS location")
plt.show()

house_box = sns.boxplot(data=facets, y='house_one',showfliers=False, color='orange').set(title="# of Housing Violations in Total Population of FIPS location (Outliers Removed)")
plt.show()
In [42]:
HV_pop = mental_health[(mental_health.house_one >= 10) & (mental_health.Total_Population <= 800)]
HV_pop.sort_values(['FIPS'])
Out[42]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
320 36005043500 2016 55 0.00 20417.0 17.0 0.2988 777.0 95.975 42.735 Bronx 2
1393 36047045000 2016 507 0.31 78438.0 33.0 0.5438 104.0 16.731 52.728 Brooklyn 3
3352 36061009400 2016 33 0.00 41667.0 17.0 0.4142 357.0 NaN 11.781 Manhattan 1
In [43]:
concern_hv = set(HV_pop['FIPS'])
#print(concern_hv)

#chv = list(HV_pop['FIPS'])
#chv

out_hv = mental_health['FIPS'].isin(concern_hv)
consistency_hv = mental_health[out_hv]

consistency_hv['Year'] = pd.to_datetime(consistency_hv['Year'], format='%Y')
consistency_hv.set_index('Year',inplace=True)


consistency_hv.groupby('FIPS')['Housing_Violations_Per_1000'].plot(legend=True,marker='o',grid=True,figsize=(5,5))
plt.title('Housing Violations Per 1000 Outlier Consistency')
plt.ylabel('Housing Violations Per 1000')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
#consistency_hv
plt.show()

consistency_hv.groupby('FIPS')['Total_Population'].plot(legend=True,marker='o',grid=True,figsize=(5,5))
plt.title('Total Population Compare to Housing Violation Per 1000')
plt.ylabel('Total Population')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
plt.show()
In [44]:
H = mental_health[(mental_health.FIPS == 36005043500) | (mental_health.FIPS == 36047045000) | (mental_health.FIPS == 36061009400)]
H.sort_values(['FIPS'])
Out[44]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
320 36005043500 2016 55 0.00 20417.0 17.0 0.2988 777.0000 95.975 42.735000 Bronx 2
656 36005043500 2017 52 0.00 42750.0 182.0 0.2455 0.0059 9.464 0.000307 Bronx 2
992 36005043500 2018 62 0.13 55181.5 17.0 0.2455 0.0059 114.018 0.000366 Bronx 2
1393 36047045000 2016 507 0.31 78438.0 33.0 0.5438 104.0000 16.731 52.728000 Brooklyn 3
2144 36047045000 2017 498 0.47 29659.0 15.0 0.6045 0.0000 7.470 0.000000 Brooklyn 3
2895 36047045000 2018 514 0.47 45938.0 47.0 0.6045 0.0000 24.158 0.000000 Brooklyn 3
3352 36061009400 2016 33 0.00 41667.0 17.0 0.4142 357.0000 NaN 11.781000 Manhattan 1
3634 36061009400 2017 15 0.00 55181.5 116.0 0.4154 0.0316 1.740 0.000474 Manhattan 1
3917 36061009400 2018 44 0.00 199167.0 17.0 0.4154 0.0316 108.988 0.001390 Manhattan 1
In [45]:
house_pop = mental_health[(mental_health.Housing_Violations_Per_1000 >= 1000)]
house_pop
Out[45]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
30 36005005600 2016 2391 0.40 17708.0 14.0 0.5665 1067.0 33.474 2551.197 Bronx 2
90 36005015700 2016 3761 0.34 31648.0 13.0 0.4638 1593.0 48.893 5991.273 Bronx 2
150 36005022500 2016 8071 0.33 31309.0 14.0 0.3937 1158.0 112.994 9346.218 Bronx 2
151 36005022701 2016 5981 0.32 29000.0 8.0 0.4122 1004.0 47.848 6004.924 Bronx 2
165 36005023702 2016 1345 0.34 28824.0 40.0 0.4778 1215.0 53.800 1634.175 Bronx 2
180 36005025100 2016 6246 0.30 32724.0 13.0 0.4415 1158.0 81.198 7232.868 Bronx 2
190 36005026500 2016 6914 0.27 33284.0 21.0 0.4828 1458.0 145.194 10080.612 Bronx 2
195 36005026900 2016 3318 0.29 34750.0 14.0 0.4269 1098.0 46.452 3643.164 Bronx 2
196 36005027300 2016 8429 0.30 36656.0 6.0 0.4270 1291.0 50.574 10881.839 Bronx 2
262 36005037300 2016 5096 0.34 30365.0 20.0 0.4164 1056.0 101.920 5381.376 Bronx 2
332 36005046201 2016 28827 0.08 44856.0 10.0 0.4250 2714.0 288.270 78236.478 Bronx 2
1010 36047000200 2016 1561 0.41 35500.0 22.0 0.3917 1876.0 34.342 2928.436 Brooklyn 3
1592 36047078200 2016 4422 0.14 48807.0 8.0 0.4122 1056.0 35.376 4669.632 Brooklyn 3
1597 36047079200 2016 3864 0.32 33932.0 16.0 0.4582 1098.0 61.824 4242.672 Brooklyn 3
1638 36047087600 2016 1630 0.12 39784.0 20.0 0.4721 1189.0 32.600 1938.070 Brooklyn 3
3528 36061027700 2016 7877 0.36 28302.0 6.0 0.4694 1029.0 47.262 8105.433 Manhattan 1
4336 36081033402 2016 12877 0.14 40736.0 10.0 0.4122 1161.0 128.770 14950.197 Queens 4
4517 36081062000 2016 1353 0.03 89621.0 5.0 0.3211 2555.0 6.765 3456.915 Queens 4
4594 36081077905 2016 2475 0.10 63776.0 5.0 0.4193 1067.0 12.375 2640.825 Queens 4
4718 36081129102 2016 3367 0.06 97031.0 6.0 0.4380 1636.0 20.202 5508.412 Queens 4
6072 36085002700 2016 2362 0.38 25957.0 11.0 0.5163 1039.0 25.982 2454.118 Staten Island 5
6106 36085013302 2016 3672 0.22 55052.0 16.0 0.4612 1456.0 58.752 5346.432 Staten Island 5
In [46]:
concern_house = set(house_pop['FIPS'])
#print(concern_house)

#chouse = list(house_pop['FIPS'])
#chouse

out_house = mental_health['FIPS'].isin(concern_house)
consistency_house = mental_health[out_house]

consistency_house['Year'] = pd.to_datetime(consistency_house['Year'], format='%Y')
consistency_house.set_index('Year',inplace=True)


consistency_house.groupby('FIPS')['Housing_Violations_Per_1000'].plot(legend=True,marker='o',grid=True,figsize=(10,10))
plt.title('Housing Violations >1000 Per 1000 Outlier Consistency')
plt.ylabel('Housing Violations Per 1000')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
#consistency_house
plt.show()

consistency_house.groupby('FIPS')['Total_Population'].plot(legend=True,marker='o',grid=True,figsize=(10,10))
plt.title('Total Population Compare to Housing Violation >1000 Per 1000 ')
plt.ylabel('Total Population')
plt.xlim([datetime.date(2015,1,1),datetime.date(2019,1,1)])
plt.show()
In [47]:
sns.boxplot(mental_health, y='Housing_Violations_Per_1000', color='orange')
plt.show()

sns.histplot(mental_health, x='Housing_Violations_Per_1000', bins=10, color='orange')
plt.show()
In [48]:
mental_health['Housing_Violations_Per_1000'] = mental_health['Housing_Violations_Per_1000'].fillna(mental_health['Housing_Violations_Per_1000'].median())

Descriptive Statistics and Null check after clean and imputation.¶

In [49]:
mental_health[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000","fel_one","house_one"]].describe()
Out[49]:
Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one
count 6378.000000 6378.000000 6378.000000 6378.000000 6378.000000 6378.000000 6364.000000 5725.000000
mean 3957.852305 0.195378 59634.980401 40.029163 0.453794 64.475070 167.418498 310.986407
std 2194.124914 0.131610 29252.545931 72.268327 0.064482 170.943849 346.922128 1349.885442
min 4.000000 0.000000 9327.000000 0.000000 0.125100 0.000000 0.000000 0.000000
25% 2428.500000 0.090000 39626.750000 9.000000 0.409850 0.121100 28.734000 0.311389
50% 3591.000000 0.170000 55181.500000 17.000000 0.449000 0.420500 60.586500 1.776917
75% 4997.750000 0.270000 73828.250000 42.000000 0.490600 8.069825 144.298500 129.920000
max 29256.000000 1.000000 250000.000000 1944.000000 0.690500 2714.000000 9661.284000 78236.478000
In [50]:
medianfill_mental = mental_health.isnull().sum()
medianfill_mental
Out[50]:
FIPS                             0
Year                             0
Total_Population                 0
Poverty_Rate                     0
MED_HH_Income                    0
Felony_Per_1000                  0
GINI                             0
Housing_Violations_Per_1000      0
fel_one                         14
house_one                      653
City                             0
boro_code                        0
dtype: int64

Boroughs Reference¶

boroughs-of-new-york-city-map.jpg

Borough boundary map to be used as base for geo heatmaps¶

In [51]:
ax = newyorkborough['geometry'].boundary.plot(figsize=(5,5))

Totals by Borough¶

In [52]:
map_mental = pd.merge(mental_health, newyorkborough, on = 'boro_code')
map_mental
Out[52]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code boro_name shape_area shape_leng geometry
0 36005000100 2016 8430 0.17 55181.5 40.0 0.4490 0.4205 337.200 NaN Bronx 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ...
1 36005000200 2016 5095 0.19 74837.0 8.0 0.4024 135.0000 40.760 687.825000 Bronx 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ...
2 36005000400 2016 5572 0.12 77991.0 10.0 0.3659 145.0000 55.720 807.940000 Bronx 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ...
3 36005001600 2016 5412 0.19 32354.0 9.0 0.4898 31.0000 48.708 167.772000 Bronx 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ...
4 36005001900 2016 2569 0.45 34635.0 43.0 0.4420 178.0000 110.467 457.282000 Bronx 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ...
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
6373 36085030301 2018 4897 0.09 65469.0 17.0 0.4005 0.5237 83.249 2.564559 Staten Island 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ...
6374 36085030302 2018 6142 0.15 84630.0 9.0 0.4038 1.3825 55.278 8.491315 Staten Island 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ...
6375 36085031901 2018 2564 0.37 23098.0 35.0 0.5083 1.1520 89.740 2.953728 Staten Island 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ...
6376 36085031902 2018 4480 0.41 23289.0 15.0 0.5671 12.0023 67.200 53.770304 Staten Island 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ...
6377 36085032300 2018 1218 0.23 64250.0 77.0 0.5786 1.0683 93.786 1.301189 Staten Island 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ...

6378 rows × 16 columns

In [53]:
sumbyboro = map_mental[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000",'City']].groupby('City').sum()
sumbyboro['boro_code'] = [2,3,1,4,5]
sumbyboro

newyorkborough.sort_values('boro_name',inplace=True)
newyorkborough
sumboro = pd.merge(newyorkborough, sumbyboro, on='boro_code')
sumboro
#print("Type of DataFrame : ", type(sumboro), sumboro.shape[0])
#print("Type of DataFrame : ", type(newyorkborough), newyorkborough.shape[0])
Out[53]:
boro_code boro_name shape_area shape_leng geometry Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000
0 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ... 4278357 295.64 40059599.0 48986.0 456.0707 109948.1328
1 3 Brooklyn 1.934143e+09 728197.541089 MULTIPOLYGON (((-73.86327 40.58388, -73.86381 ... 7772912 482.88 123307368.5 77897.0 1039.2900 172008.8464
2 1 Manhattan 6.365208e+08 357564.316391 MULTIPOLYGON (((-74.01093 40.68449, -74.01193 ... 4882878 151.01 71205121.0 65013.0 441.1521 44208.1009
3 4 Queens 3.041419e+09 888199.731579 MULTIPOLYGON (((-73.82645 40.59053, -73.82642 ... 6891708 273.67 122250607.0 54413.0 818.5827 71468.8388
4 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ... 1417327 42.92 23529209.5 8997.0 139.2043 13588.0774
In [54]:
fig, ax = plt.subplots(figsize=(5,5))
sumboro.plot(column='Total_Population', ax=ax, cmap='rainbow', legend= True, legend_kwds={"label": "Total Population Sum"})
ax.set_title("Total Population 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
sumboro.plot(column='Poverty_Rate', ax=ax, cmap='rainbow', legend= True, legend_kwds={"label": "Poverty Rate Sum"})
ax.set_title("Poverty Rate 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
sumboro.plot(column='MED_HH_Income', ax=ax, cmap='rainbow', legend= True, legend_kwds={"label": "Median Household Income Sum"})
ax.set_title("Median Household Income 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
sumboro.plot(column='Felony_Per_1000', ax=ax, cmap='rainbow', legend= True, legend_kwds={"label": "Felony Per 1000 Sum"})
ax.set_title("Felony Per 1000 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
sumboro.plot(column='GINI', ax=ax, cmap='rainbow', legend= True, legend_kwds={"label": "GINI Sum"})
ax.set_title("GINI 2016-2018")

fig, ax = plt.subplots(figsize=(5,5))
sumboro.plot(column='Housing_Violations_Per_1000', ax=ax, cmap='rainbow', legend= True, legend_kwds={"label": "Housing Violations Sum"})
ax.set_title("Housing Violations Per 1000 2016-2018")
Out[54]:
Text(0.5, 1.0, 'Housing Violations Per 1000 2016-2018')

Percentage¶

In [55]:
percentbyboro = pd.DataFrame()
percentbyboro['Percent_Total_Pop'] = (sumbyboro['Total_Population']/sumbyboro['Total_Population'].sum())*100
percentbyboro['Percent_Poverty_Rate'] = (sumbyboro['Poverty_Rate']/sumbyboro['Poverty_Rate'].sum())*100
percentbyboro['Percent_Median_Income'] = (sumbyboro['MED_HH_Income']/sumbyboro['MED_HH_Income'].sum())*100
percentbyboro['Percent_Felony'] = (sumbyboro['Felony_Per_1000']/sumbyboro['Felony_Per_1000'].sum())*100
percentbyboro['Percent_GINI'] = (sumbyboro['GINI']/sumbyboro['GINI'].sum())*100
percentbyboro['Percent_Housing'] = (sumbyboro['Housing_Violations_Per_1000']/sumbyboro['Housing_Violations_Per_1000'].sum())*100
percentbyboro['boro_code'] = [2,3,1,4,5]
percentbyboro

percentboro = pd.merge(newyorkborough, percentbyboro, on='boro_code')
percentboro

fig, ax = plt.subplots(figsize=(5,5))
percentboro.plot(column='Percent_Total_Pop', ax=ax, cmap='plasma', legend= True, legend_kwds={"label": "Total Population %"})
ax.set_title("Total Population % 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
percentboro.plot(column='Percent_Poverty_Rate', ax=ax, cmap='plasma', legend= True, legend_kwds={"label": "Poverty Rate %"})
ax.set_title("Poverty Rate % 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
percentboro.plot(column='Percent_Median_Income', ax=ax, cmap='plasma', legend= True, legend_kwds={"label": "Median Household Income %"})
ax.set_title("Median Household Income % 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
percentboro.plot(column='Percent_Felony', ax=ax, cmap='plasma', legend= True, legend_kwds={"label": "Felony Per 1000 %"})
ax.set_title("Felony Per 1000 % 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
percentboro.plot(column='Percent_GINI', ax=ax, cmap='plasma', legend= True, legend_kwds={"label": "GINI %"})
ax.set_title("GINI % 2016-2018")

fig, ax = plt.subplots(figsize=(5,5))
percentboro.plot(column='Percent_Housing', ax=ax, cmap='plasma', legend= True, legend_kwds={"label": "Housing Violations %"})
ax.set_title("Housing Violations Per 1000 % 2016-2018")
Out[55]:
Text(0.5, 1.0, 'Housing Violations Per 1000 % 2016-2018')

Averages by Borough¶

In [56]:
avgbyboro = map_mental[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000",'City']].groupby('City').mean()
avgbyboro['boro_code'] = [2,3,1,4,5]
avgbyboro

newyorkborough.sort_values('boro_name',inplace=True)
newyorkborough
avgboro = pd.merge(newyorkborough, avgbyboro, on='boro_code')
avgboro
#print("Type of DataFrame : ", type(avgboro), avgboro.shape[0])
#print("Type of DataFrame : ", type(newyorkborough), newyorkborough.shape[0])


fig, ax = plt.subplots(figsize=(5,5))
avgboro.plot(column='Total_Population', ax=ax, cmap='viridis', legend= True, legend_kwds={"label": "Total Population Average"})
ax.set_title("Total Population Average 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
avgboro.plot(column='Poverty_Rate', ax=ax, cmap='viridis', legend= True, legend_kwds={"label": "Poverty Rate Average"})
ax.set_title("Poverty Rate Average 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
avgboro.plot(column='MED_HH_Income', ax=ax, cmap='viridis', legend= True, legend_kwds={"label": "Median Household Income Average"})
ax.set_title("Median Household Income Average 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
avgboro.plot(column='Felony_Per_1000', ax=ax, cmap='viridis', legend= True, legend_kwds={"label": "Felony Per 1000 Average"})
ax.set_title("Felony Per 1000 Average 2016-2018")


fig, ax = plt.subplots(figsize=(5,5))
avgboro.plot(column='GINI', ax=ax, cmap='viridis', legend= True, legend_kwds={"label": "GINI Average"})
ax.set_title("GINI Average 2016-2018")

fig, ax = plt.subplots(figsize=(5,5))
avgboro.plot(column='Housing_Violations_Per_1000', ax=ax, cmap='viridis', legend= True, legend_kwds={"label": "Housing Violations Average"})
ax.set_title("Housing Violations Per 1000 Average 2016-2018")
Out[56]:
Text(0.5, 1.0, 'Housing Violations Per 1000 Average 2016-2018')

Yearly Breakdown¶

2016 Totals¶

In [57]:
year16.head()
Out[57]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
0 36005000100 2016 8430 NaN NaN 40.0 NaN NaN 337.200 NaN Bronx 2
1 36005000200 2016 5095 0.19 74837.0 8.0 0.4024 135.0 40.760 687.825 Bronx 2
2 36005000400 2016 5572 0.12 77991.0 10.0 0.3659 145.0 55.720 807.940 Bronx 2
3 36005001600 2016 5412 0.19 32354.0 9.0 0.4898 31.0 48.708 167.772 Bronx 2
4 36005001900 2016 2569 0.45 34635.0 43.0 0.4420 178.0 110.467 457.282 Bronx 2
In [58]:
sumby16 = year16[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000",'City']].groupby('City').sum()
sumby16['boro_code'] = [2,3,1,4,5]
sumby16

sum16 = pd.merge(newyorkborough, sumby16, on='boro_code')
sum16
#print("Type of DataFrame : ", type(sum16), sum16.shape[0])
Out[58]:
boro_code boro_name shape_area shape_leng geometry Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000
0 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ... 1413566 99.07 13070037.0 8054.0 149.5011 108968.0
1 3 Brooklyn 1.934143e+09 728197.541089 MULTIPOLYGON (((-73.86327 40.58388, -73.86381 ... 2570801 162.39 39741318.0 9720.0 342.4496 171454.0
2 1 Manhattan 6.365208e+08 357564.316391 MULTIPOLYGON (((-74.01093 40.68449, -74.01193 ... 1618398 50.11 22990893.0 11429.0 145.5053 43826.0
3 4 Queens 3.041419e+09 888199.731579 MULTIPOLYGON (((-73.82645 40.59053, -73.82642 ... 2280602 92.81 39788374.0 6167.0 268.5323 71142.0
4 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ... 471522 13.80 7802279.0 756.0 45.7055 13291.0
In [59]:
fig, ax = plt.subplots(figsize=(5,5))
sum16.plot(column='Total_Population', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Total Population Sum"})
ax.set_title("Total Population 2016")


fig, ax = plt.subplots(figsize=(5,5))
sum16.plot(column='Poverty_Rate', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Poverty Rate Sum"})
ax.set_title("Poverty Rate 2016")


fig, ax = plt.subplots(figsize=(5,5))
sum16.plot(column='MED_HH_Income', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Median Household Income Sum"})
ax.set_title("Median Household Income 2016")


fig, ax = plt.subplots(figsize=(5,5))
sum16.plot(column='Felony_Per_1000', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Felony Per 1000 Sum"})
ax.set_title("Felony Per 1000 2016")


fig, ax = plt.subplots(figsize=(5,5))
sum16.plot(column='GINI', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "GINI Sum"})
ax.set_title("GINI 2016")

fig, ax = plt.subplots(figsize=(5,5))
sum16.plot(column='Housing_Violations_Per_1000', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Housing Violations Sum"})
ax.set_title("Housing Violations Per 1000 2016")
Out[59]:
Text(0.5, 1.0, 'Housing Violations Per 1000 2016')

2016 Percentages¶

In [60]:
percentby2016 = pd.DataFrame()
percentby2016['Percent_Total_Pop'] = (sumby16['Total_Population']/sumby16['Total_Population'].sum())*100
percentby2016['Percent_Poverty_Rate'] = (sumby16['Poverty_Rate']/sumby16['Poverty_Rate'].sum())*100
percentby2016['Percent_Median_Income'] = (sumby16['MED_HH_Income']/sumby16['MED_HH_Income'].sum())*100
percentby2016['Percent_Felony'] = (sumby16['Felony_Per_1000']/sumby16['Felony_Per_1000'].sum())*100
percentby2016['Percent_GINI'] = (sumby16['GINI']/sumby16['GINI'].sum())*100
percentby2016['Percent_Housing'] = (sumby16['Housing_Violations_Per_1000']/sumby16['Housing_Violations_Per_1000'].sum())*100
percentby2016['boro_code'] = [2,3,1,4,5]
display(percentby2016)

percent2016 = pd.merge(newyorkborough, percentby2016, on='boro_code')
percent2016

fig, ax = plt.subplots(figsize=(5,5))
percent2016.plot(column='Percent_Total_Pop', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Total Population %"})
ax.set_title("Total Population % 2016")


fig, ax = plt.subplots(figsize=(5,5))
percent2016.plot(column='Percent_Poverty_Rate', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Poverty Rate %"})
ax.set_title("Poverty Rate % 2016")


fig, ax = plt.subplots(figsize=(5,5))
percent2016.plot(column='Percent_Median_Income', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Median Household Income %"})
ax.set_title("Median Household Income % 2016")


fig, ax = plt.subplots(figsize=(5,5))
percent2016.plot(column='Percent_Felony', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Felony Per 1000 %"})
ax.set_title("Felony Per 1000 % 2016")


fig, ax = plt.subplots(figsize=(5,5))
percent2016.plot(column='Percent_GINI', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "GINI %"})
ax.set_title("GINI % 2016")

fig, ax = plt.subplots(figsize=(5,5))
percent2016.plot(column='Percent_Housing', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Housing Violations %"})
ax.set_title("Housing Violations Per 1000 % 2016")
Percent_Total_Pop Percent_Poverty_Rate Percent_Median_Income Percent_Felony Percent_GINI Percent_Housing boro_code
City
Bronx 16.919028 23.690755 10.592211 22.294193 15.708950 26.663339 2
Brooklyn 30.770020 38.832560 32.207135 26.905830 35.983170 41.953015 3
Manhattan 19.370670 11.982878 18.632266 31.636494 15.289088 10.723767 1
Queens 27.296616 22.193792 32.245270 17.070808 28.216250 17.407709 4
Staten Island 5.643666 3.300014 6.323118 2.092676 4.802543 3.252170 5
Out[60]:
Text(0.5, 1.0, 'Housing Violations Per 1000 % 2016')

2016 Averages¶

In [61]:
avgby2016 = year16[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000",'City']].groupby('City').mean()
avgby2016['boro_code'] = [2,3,1,4,5]
display(avgby2016)

avg2016 = pd.merge(newyorkborough, avgby2016, on='boro_code')
avg2016
#print("Type of DataFrame : ", type(avg2016), avg2016.shape[0])


fig, ax = plt.subplots(figsize=(5,5))
avg2016.plot(column='Total_Population', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Total Population Average"})
ax.set_title("Total Population Average 2016")

avg2016['coords'] = avg2016['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2016['coords'] = [coords[0] for coords in avg2016['coords']]
for idx, row in avg2016.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2016.plot(column='Poverty_Rate', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Poverty Rate Average"})
ax.set_title("Poverty Rate Average 2016")

avg2016['coords'] = avg2016['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2016['coords'] = [coords[0] for coords in avg2016['coords']]
for idx, row in avg2016.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2016.plot(column='MED_HH_Income', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Median Household Income Average"})
ax.set_title("Median Household Income Average 2016")

avg2016['coords'] = avg2016['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2016['coords'] = [coords[0] for coords in avg2016['coords']]
for idx, row in avg2016.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2016.plot(column='Felony_Per_1000', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Felony Per 1000 Average"})
ax.set_title("Felony Per 1000 Average 2016")

avg2016['coords'] = avg2016['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2016['coords'] = [coords[0] for coords in avg2016['coords']]
for idx, row in avg2016.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2016.plot(column='GINI', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "GINI Average"})
ax.set_title("GINI Average 2016")

avg2016['coords'] = avg2016['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2016['coords'] = [coords[0] for coords in avg2016['coords']]
for idx, row in avg2016.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2016.plot(column='Housing_Violations_Per_1000', ax=ax, cmap='autumn', legend= True, legend_kwds={"label": "Housing Violations Average"})
ax.set_title("Housing Violations Per 1000 Average 2016")

avg2016['coords'] = avg2016['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2016['coords'] = [coords[0] for coords in avg2016['coords']]
for idx, row in avg2016.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')
Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 boro_code
City
Bronx 4194.557864 0.294851 39249.360360 23.970238 0.448952 408.119850 2
Brooklyn 3423.170439 0.216520 53130.104278 12.977303 0.457820 272.581876 3
Manhattan 5739.000000 0.178327 82404.634409 40.964158 0.521524 188.094421 1
Queens 3503.228879 0.142785 61879.275272 9.576087 0.416976 174.367647 4
Staten Island 4365.944444 0.127778 72918.495327 7.065421 0.427154 221.516667 5

2017 Totals¶

In [62]:
year17.head()
Out[62]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
337 36005000100 2017 7703 NaN NaN 955.0 NaN NaN 7356.365 NaN Bronx 2
338 36005000200 2017 5403 0.20 72034.0 71.0 0.3903 0.0967 383.613 0.522470 Bronx 2
339 36005000400 2017 5915 0.13 74836.0 77.0 0.3644 0.3370 455.455 1.993355 Bronx 2
340 36005001600 2017 5879 0.26 32312.0 86.0 0.5018 0.1817 505.594 1.068214 Bronx 2
341 36005001900 2017 2591 0.37 37936.0 193.0 0.4260 0.5216 500.063 1.351466 Bronx 2
In [63]:
sumby17 = year17[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000",'City']].groupby('City').sum()
sumby17['boro_code'] = [2,3,1,4,5]
sumby17

sum17 = pd.merge(newyorkborough, sumby17, on='boro_code')
sum17
#print("Type of DataFrame : ", type(sum17), sum17.shape[0])
Out[63]:
boro_code boro_name shape_area shape_leng geometry Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000
0 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ... 1428357 98.76 12933386.0 35442.0 151.9378 474.0874
1 3 Brooklyn 1.934143e+09 728197.541089 MULTIPOLYGON (((-73.86327 40.58388, -73.86381 ... 2595259 161.53 40590155.0 53766.0 346.3997 244.2037
2 1 Manhattan 6.365208e+08 357564.316391 MULTIPOLYGON (((-74.01093 40.68449, -74.01193 ... 1629507 50.54 23421786.0 42011.0 145.8029 177.8047
3 4 Queens 3.041419e+09 888199.731579 MULTIPOLYGON (((-73.82645 40.59053, -73.82642 ... 2301139 91.70 40020276.0 36917.0 271.4332 102.0264
4 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ... 472481 14.25 7706272.0 6048.0 46.5249 135.0827
In [64]:
fig, ax = plt.subplots(figsize=(5,5))
sum17.plot(column='Total_Population', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Total Population Sum"})
ax.set_title("Total Population 2017")


fig, ax = plt.subplots(figsize=(5,5))
sum17.plot(column='Poverty_Rate', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Poverty Rate Sum"})
ax.set_title("Poverty Rate 2017")


fig, ax = plt.subplots(figsize=(5,5))
sum17.plot(column='MED_HH_Income', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Median Household Income Sum"})
ax.set_title("Median Household Income 2017")


fig, ax = plt.subplots(figsize=(5,5))
sum17.plot(column='Felony_Per_1000', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Felony Per 1000 Sum"})
ax.set_title("Felony Per 1000 2017")


fig, ax = plt.subplots(figsize=(5,5))
sum17.plot(column='GINI', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "GINI Sum"})
ax.set_title("GINI 2017")

fig, ax = plt.subplots(figsize=(5,5))
sum17.plot(column='Housing_Violations_Per_1000', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Housing Violations Sum"})
ax.set_title("Housing Violations Per 1000 2017")
Out[64]:
Text(0.5, 1.0, 'Housing Violations Per 1000 2017')

2017 Percentages¶

In [65]:
percentby2017 = pd.DataFrame()
percentby2017['Percent_Total_Pop'] = (sumby17['Total_Population']/sumby17['Total_Population'].sum())*100
percentby2017['Percent_Poverty_Rate'] = (sumby17['Poverty_Rate']/sumby17['Poverty_Rate'].sum())*100
percentby2017['Percent_Median_Income'] = (sumby17['MED_HH_Income']/sumby17['MED_HH_Income'].sum())*100
percentby2017['Percent_Felony'] = (sumby17['Felony_Per_1000']/sumby17['Felony_Per_1000'].sum())*100
percentby2017['Percent_GINI'] = (sumby17['GINI']/sumby17['GINI'].sum())*100
percentby2017['Percent_Housing'] = (sumby17['Housing_Violations_Per_1000']/sumby17['Housing_Violations_Per_1000'].sum())*100
percentby2017['boro_code'] = [2,3,1,4,5]
display(percentby2017)

percent2017 = pd.merge(newyorkborough, percentby2017, on='boro_code')
percent2017

fig, ax = plt.subplots(figsize=(5,5))
percent2017.plot(column='Percent_Total_Pop', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Total Population %"})
ax.set_title("Total Population % 2017")


fig, ax = plt.subplots(figsize=(5,5))
percent2017.plot(column='Percent_Poverty_Rate', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Poverty Rate %"})
ax.set_title("Poverty Rate % 2017")


fig, ax = plt.subplots(figsize=(5,5))
percent2017.plot(column='Percent_Median_Income', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Median Household Income %"})
ax.set_title("Median Household Income % 2017")


fig, ax = plt.subplots(figsize=(5,5))
percent2017.plot(column='Percent_Felony', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Felony Per 1000 %"})
ax.set_title("Felony Per 1000 % 2017")


fig, ax = plt.subplots(figsize=(5,5))
percent2017.plot(column='Percent_GINI', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "GINI %"})
ax.set_title("GINI % 2017")

fig, ax = plt.subplots(figsize=(5,5))
percent2017.plot(column='Percent_Housing', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Housing Violations %"})
ax.set_title("Housing Violations Per 1000 % 2017")
Percent_Total_Pop Percent_Poverty_Rate Percent_Median_Income Percent_Felony Percent_GINI Percent_Housing boro_code
City
Bronx 16.950286 23.695955 10.373940 20.347449 15.792333 41.835982 2
Brooklyn 30.797890 38.756658 32.557588 30.867359 36.004598 21.549827 3
Manhattan 19.337329 12.126302 18.786744 24.118748 15.154675 15.690428 1
Queens 27.307573 22.002015 32.100485 21.194254 28.212621 9.003350 4
Staten Island 5.606923 3.419070 6.181243 3.472190 4.835773 11.920413 5
Out[65]:
Text(0.5, 1.0, 'Housing Violations Per 1000 % 2017')

2017 Averages¶

In [66]:
avgby2017 = year17[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000",'City']].groupby('City').mean()
avgby2017['boro_code'] = [2,3,1,4,5]
display(avgby2017)

avg2017 = pd.merge(newyorkborough, avgby2017, on='boro_code')
avg2017
#print("Type of DataFrame : ", type(avg2016), avg2016.shape[0])



fig, ax = plt.subplots(figsize=(5,5))
avg2017.plot(column='Total_Population', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Total Population Average"})
ax.set_title("Total Population Average 2017")

avg2017['coords'] = avg2017['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2017['coords'] = [coords[0] for coords in avg2017['coords']]
for idx, row in avg2017.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2017.plot(column='Poverty_Rate', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Poverty Rate Average"})
ax.set_title("Poverty Rate Average 2017")

avg2017['coords'] = avg2017['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2017['coords'] = [coords[0] for coords in avg2017['coords']]
for idx, row in avg2017.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2017.plot(column='MED_HH_Income', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Median Household Income Average"})
ax.set_title("Median Household Income Average 2017")

avg2017['coords'] = avg2017['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2017['coords'] = [coords[0] for coords in avg2017['coords']]
for idx, row in avg2017.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2017.plot(column='Felony_Per_1000', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Felony Per 1000 Average"})
ax.set_title("Felony Per 1000 Average 2017")

avg2017['coords'] = avg2017['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2017['coords'] = [coords[0] for coords in avg2017['coords']]
for idx, row in avg2017.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2017.plot(column='GINI', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "GINI Average"})
ax.set_title("GINI Average 2017")

avg2017['coords'] = avg2017['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2017['coords'] = [coords[0] for coords in avg2017['coords']]
for idx, row in avg2017.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2017.plot(column='Housing_Violations_Per_1000', ax=ax, cmap='spring', legend= True, legend_kwds={"label": "Housing Violations Average"})
ax.set_title("Housing Violations Per 1000 Average 2017")

avg2017['coords'] = avg2017['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2017['coords'] = [coords[0] for coords in avg2017['coords']]
for idx, row in avg2017.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')
Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 boro_code
City
Bronx 4251.062500 0.294806 39311.203647 105.482143 0.456270 1.432288 2
Brooklyn 3455.737683 0.215373 54264.913102 71.592543 0.463101 0.333156 3
Manhattan 5757.975265 0.179220 84555.184116 148.448763 0.522591 0.644220 1
Queens 3540.213846 0.141077 62336.878505 56.970679 0.420827 0.163242 4
Staten Island 4374.824074 0.131944 72021.233645 56.000000 0.430786 1.350827 5

2018 Totals¶

In [67]:
year18.head()
Out[67]:
FIPS Year Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 fel_one house_one City boro_code
673 36005000100 2018 7503 NaN NaN 89.0 NaN NaN 667.767 NaN Bronx 2
674 36005000200 2018 5251 0.16 70893.0 10.0 0.3903 0.0967 52.510 0.507772 Bronx 2
675 36005000400 2018 5980 0.12 76667.0 12.0 0.3644 0.3370 71.760 2.015260 Bronx 2
676 36005001600 2018 6056 0.29 31540.0 11.0 0.5018 0.1817 66.616 1.100375 Bronx 2
677 36005001900 2018 2682 0.35 39130.0 59.0 0.4260 0.5216 158.238 1.398931 Bronx 2
In [68]:
sumby18 = year18[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000",'City']].groupby('City').sum()
sumby18['boro_code'] = [2,3,1,4,5]
sumby18

sum18 = pd.merge(newyorkborough, sumby18, on='boro_code')
sum18
#print("Type of DataFrame : ", type(sum18), sum18.shape[0])
Out[68]:
boro_code boro_name shape_area shape_leng geometry Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000
0 2 Bronx 1.187175e+09 463179.772813 MULTIPOLYGON (((-73.89681 40.79581, -73.89694 ... 1436785 97.87 13283635.0 9636.0 151.9378 474.0874
1 3 Brooklyn 1.934143e+09 728197.541089 MULTIPOLYGON (((-73.86327 40.58388, -73.86381 ... 2606852 158.45 42479262.0 16360.0 346.3997 244.2037
2 1 Manhattan 6.365208e+08 357564.316391 MULTIPOLYGON (((-74.01093 40.68449, -74.01193 ... 1634989 49.85 24130264.0 72721.0 145.8029 177.8047
3 4 Queens 3.041419e+09 888199.731579 MULTIPOLYGON (((-73.82645 40.59053, -73.82642 ... 2310011 89.16 41117601.0 18926.0 271.4332 102.0264
4 5 Staten Island 1.623621e+09 325917.353702 MULTIPOLYGON (((-74.05051 40.56642, -74.05047 ... 473324 14.87 7855114.0 2176.0 46.5249 135.0827
In [69]:
fig, ax = plt.subplots(figsize=(5,5))
sum18.plot(column='Total_Population', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Total Population Sum"})
ax.set_title("Total Population 2018")


fig, ax = plt.subplots(figsize=(5,5))
sum18.plot(column='Poverty_Rate', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Poverty Rate Sum"})
ax.set_title("Poverty Rate 2018")


fig, ax = plt.subplots(figsize=(5,5))
sum18.plot(column='MED_HH_Income', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Median Household Income Sum"})
ax.set_title("Median Household Income 2018")


fig, ax = plt.subplots(figsize=(5,5))
sum18.plot(column='Felony_Per_1000', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Felony Per 1000 Sum"})
ax.set_title("Felony Per 1000 2018")


fig, ax = plt.subplots(figsize=(5,5))
sum18.plot(column='GINI', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "GINI Sum"})
ax.set_title("GINI 2018")

fig, ax = plt.subplots(figsize=(5,5))
sum18.plot(column='Housing_Violations_Per_1000', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Housing Violations Sum"})
ax.set_title("Housing Violations Per 1000 2018")
Out[69]:
Text(0.5, 1.0, 'Housing Violations Per 1000 2018')

2018 Percentages¶

In [70]:
percentby2018 = pd.DataFrame()
percentby2018['Percent_Total_Pop'] = (sumby18['Total_Population']/sumby18['Total_Population'].sum())*100
percentby2018['Percent_Poverty_Rate'] = (sumby18['Poverty_Rate']/sumby18['Poverty_Rate'].sum())*100
percentby2018['Percent_Median_Income'] = (sumby18['MED_HH_Income']/sumby18['MED_HH_Income'].sum())*100
percentby2018['Percent_Felony'] = (sumby18['Felony_Per_1000']/sumby18['Felony_Per_1000'].sum())*100
percentby2018['Percent_GINI'] = (sumby18['GINI']/sumby18['GINI'].sum())*100
percentby2018['Percent_Housing'] = (sumby18['Housing_Violations_Per_1000']/sumby17['Housing_Violations_Per_1000'].sum())*100
percentby2018['boro_code'] = [2,3,1,4,5]
percentby2018

percent2018 = pd.merge(newyorkborough, percentby2018, on='boro_code')
percent2018

fig, ax = plt.subplots(figsize=(5,5))
percent2018.plot(column='Percent_Total_Pop', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Total Population %"})
ax.set_title("Total Population % 2018")


fig, ax = plt.subplots(figsize=(5,5))
percent2018.plot(column='Percent_Poverty_Rate', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Poverty Rate %"})
ax.set_title("Poverty Rate % 2018")


fig, ax = plt.subplots(figsize=(5,5))
percent2018.plot(column='Percent_Median_Income', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Median Household Income %"})
ax.set_title("Median Household Income % 2018")


fig, ax = plt.subplots(figsize=(5,5))
percent2018.plot(column='Percent_Felony', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Felony Per 1000 %"})
ax.set_title("Felony Per 1000 % 2018")


fig, ax = plt.subplots(figsize=(5,5))
percent2018.plot(column='Percent_GINI', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "GINI %"})
ax.set_title("GINI % 2018")

fig, ax = plt.subplots(figsize=(5,5))
percent2018.plot(column='Percent_Housing', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Housing Violations %"})
ax.set_title("Housing Violations Per 1000 % 2018")
Out[70]:
Text(0.5, 1.0, 'Housing Violations Per 1000 % 2018')

2018 Averages¶

In [71]:
avgby2018 = year18[["Total_Population","Poverty_Rate", "MED_HH_Income","Felony_Per_1000",
               "GINI","Housing_Violations_Per_1000",'City']].groupby('City').mean()
avgby2018['boro_code'] = [2,3,1,4,5]
display(avgby2018)

avg2018 = pd.merge(newyorkborough, avgby2018, on='boro_code')
avg2018
#print("Type of DataFrame : ", type(avg2018), avg2018.shape[0])



fig, ax = plt.subplots(figsize=(5,5))
avg2018.plot(column='Total_Population', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Total Population Average"})
ax.set_title("Total Population Average 2018")

avg2018['coords'] = avg2018['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2018['coords'] = [coords[0] for coords in avg2018['coords']]
for idx, row in avg2018.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2018.plot(column='Poverty_Rate', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Poverty Rate Average"})
ax.set_title("Poverty Rate Average 2018")

avg2018['coords'] = avg2018['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2018['coords'] = [coords[0] for coords in avg2018['coords']]
for idx, row in avg2018.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2018.plot(column='MED_HH_Income', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Median Household Income Average"})
ax.set_title("Median Household Income Average 2018")

avg2018['coords'] = avg2018['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2018['coords'] = [coords[0] for coords in avg2018['coords']]
for idx, row in avg2018.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2018.plot(column='Felony_Per_1000', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Felony Per 1000 Average"})
ax.set_title("Felony Per 1000 Average 2018")

avg2018['coords'] = avg2018['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2018['coords'] = [coords[0] for coords in avg2018['coords']]
for idx, row in avg2018.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2018.plot(column='GINI', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "GINI Average"})
ax.set_title("GINI Average 2018")

avg2018['coords'] = avg2018['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2018['coords'] = [coords[0] for coords in avg2018['coords']]
for idx, row in avg2018.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')

fig, ax = plt.subplots(figsize=(5,5))
avg2018.plot(column='Housing_Violations_Per_1000', ax=ax, cmap='summer', legend= True, legend_kwds={"label": "Housing Violations Average"})
ax.set_title("Housing Violations Per 1000 Average 2018")

avg2018['coords'] = avg2018['geometry'].apply(lambda x: x.representative_point().coords[:])
avg2018['coords'] = [coords[0] for coords in avg2018['coords']]
for idx, row in avg2018.iterrows():
   plt.annotate(text=row['boro_name'], xy=row['coords'], horizontalalignment='center', color='black')
Total_Population Poverty_Rate MED_HH_Income Felony_Per_1000 GINI Housing_Violations_Per_1000 boro_code
City
Bronx 4276.145833 0.292149 40375.790274 28.678571 0.456270 1.432288 2
Brooklyn 3471.174434 0.211267 56790.457219 21.784288 0.463101 0.333156 3
Manhattan 5777.346290 0.176773 86799.510791 256.964664 0.522591 0.644220 1
Queens 3553.863077 0.137169 64146.023401 29.116923 0.420827 0.163242 4
Staten Island 4382.629630 0.137685 73412.280374 20.148148 0.430786 1.350827 5
In [ ]: